Point-Based Planning for Multi-Objective POMDPs

Authors: Diederik Marijn Roijers, Shimon Whiteson, Frans A. Oliehoek

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We show experimentally that OLSAR outperforms, both in terms of runtime and approximation quality, alternative methods and a variant of OLSAR that does not leverage reuse.
Researcher Affiliation Academia 1Informatics Institute, University of Amsterdam, The Netherlands 2Department of Computer Science, University of Liverpool, United Kingdom
Pseudocode Yes Algorithm 1: OLSAR(b0, η) and Algorithm 2: OCPerseus(A, B, w, η)
Open Source Code No The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets No The paper uses benchmark POMDP problems (Tiger, Maze20) and generates 'sampled beliefs' for its experiments. It does not provide concrete access information (link, DOI, citation) for a publicly available, pre-existing dataset that is 'trained' on in the traditional sense.
Dataset Splits No The paper mentions generating a 'reference set' for comparison, but it does not specify explicit training/validation/test dataset splits (e.g., percentages or counts of a fixed dataset) for reproducibility.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU/GPU models, memory specifications) used for running the experiments.
Software Dependencies No The paper describes the algorithms and their implementation details but does not provide specific version numbers for any software dependencies or libraries used.
Experiment Setup Yes We ran all algorithms with 100 belief points generated by random exploration, η = 1 × 10−6, and b0 set to a uniform distribution.